More on Choosing Songs
I have written several times on various aspects of choosing music to perform and/or to arrange. These have covered both technical and artistic criteria, and also given some ideas about process - not just what to look for but how to go about looking. I had an email conversation recently, though, with a chorus director who was looking to commission an arrangement that opened up an area right in the middle of the issues I have previously covered, but which I haven’t actually written about: which specific features should she advise her chorus members to look out for in a song that would mark it as suitable for a cappella arrangement?
Now, I used to dedicate a whole class to this question when I used to teach a course on Vocal Close Harmony at Birmingham Conservatoire. So my first instinct was just to dig out those notes and post ‘em up. But, four years on from when I last taught that course, and more years than that since I taught it in the format that included that session, I can find no trace of those notes. Deep sigh. So, we’ll have to do the thinking again from scratch.
Actually, the ideas my correspondent floated as possible criteria offer a pretty good starting point:
Things that I think about are whether there’s an interesting melody line, whether the bass/chords are fairly static (bad) or whether they ‘do something’ (good), whether there are any clear harmony hints there (like already using vocal harmonies, or whether you can hear instruments playing obvious harmony lines!) but I’m not sure I would clearly be able to put this into something that would be of use to the chorus.
These are indeed all good things to think about.
Possibly a more systematic approach could be to think about the song in terms of what barbershoppers are gradually edging away from calling its ‘theme’ - what is the leading musical element, the aspect of the song that sticks in your mind when all else fades?
If the most distinctive dimension is the melody or the harmony, then you’ll find that arranging it for unaccompanied voices preserves, indeed features, the most important elements of the song. If the thing that makes the song interesting is timbral - the sound colours of original vocalist and/or instrumental accompaniment, then you may find that it does not lend itself so well. Not saying it is impossible, but the arrangement is less likely to be one of those that ‘sings itself’. (I was once commissioned to produce an a cappella arrangement of Brian Eno’s ‘Golden Hours’, which was something of a challenge, though it was for studio recording with no limit placed on the number of tracks I could arrange for.)
If the main interest seems to be lyric or rhythm, then it could go either way - and which way will largely be guided by the above considerations.
Another way to think about it is to consider the performance history of a song. Even very recent music pops up in multiple cover versions on youtube, and popular music of a decade or more ago will frequently have professional as well as amateur covers to browse through. How these work tells you a lot about the general transferability of the song from medium to medium.
If a song that was originally lavishly produced still sounds great sung by a guy with an acoustic guitar, that suggests it is the portable dimensions of the song that hold its character and identity. If there are cover versions with quite different expressive feels, again, this vouches for the song’s adaptability. If every cover version is basically a Stars-in-Their-Eyes impersonation of the original, then the heart of the song may be too tied to a specific voice to sound well in a different context.
But possibly the most primal sorting method is the woodshedding test. When you hear the song, do you get an overwhelming urge to harmonise along with it? Listen to that musical instinct; it is usually right.